DIKSHA GOYAL
AI Manager
Jaipur
Summary
Meticulous Data Scientist accomplished in compiling, transforming and analyzing complex information through software. Expert in machine learning and large dataset management. Demonstrated success in identifying relationships and building solutions to business problems.
Overview
8
8
years of professional experience
46
46
years of post-secondary education
1
1
Certificate
Work History
Senior Data Scientist
Hero Moto Corp Limited
03.2022 - Current
- Battery Analytics and Predictive Maintenance:
Monitor battery performance, predict maintenance requirements, optimize energy usage, design more efficient battery systems, and offer new services to the customers. By analyzing the battery data, we can predict when maintenance or replacement of the battery pack may be required.
- Battery Remaining Useful Life:
Developing testing scenarios for on road driving and simulations in lab. Measure battery performance parameters and develop a life prediction model for Battery RUL.
- Rider Behavior Analysis :
Building profiles for individual riders using accelerometer data, Vehicle telemetry, riding style, preferred speed ranges, braking patterns etc
- Range Prediction and Range anxiety Solutions:
By understanding the rider's typical routes, riding style and energy consumption patterns, using the rider behavior data, real time sensors data, riding conditions, State of Charge, State of health, Weather etc. to improve range prediction in EV Two-Wheeler
- Future scope of Battery Secondary Life:
Exploring and establishing solutions on optimum usage of battery at lower SOH levels after Electric vehicle usage.
Assistant Manager
Tata Power Delhi Distribution Limited
06.2018 - 02.2022
Revenue Protection System:
- Unified theft detection system to identify theft prone/default meter device.
- Comprising of Data Accumulation, Data Analysis, Reporting, Manual Analysis and Alerting Layer.
- Identify losses at DT level for efficient energy auditing integrated with geographical information system.
Identification of Bill-defaulters:
- Built a bill defaulter prediction system based on consumer demographic and behavioral pattern of previous bill payments to improve the billing efficiency of every month and reduce AT&C loss level. (Accuracy ~40%).
Signature Analysis of Various Industry Usage pattern of energy:
- To identify the consumers which are deviating from their industry’s base load pattern. (They might indulge in theft activity)
Energy Consumption Forecast:
- Developed a time series model to predict future energy consumption for next one month using seasonality and trend.
Senior Executive
Tata Power Delhi Distribution Limited
07.2016 - 05.2018
Energy Meter Data Analysis:
- Identification of dishonest use of energy connections as per consumer's usage pattern
- Data Study of probable theft prone connections i.e change in billed units, voltage and current values etc
- Developed in-house logics for theft identification of high revenue base consumers
Advanced Metering Infrastructure (AMI) Implementation (Using RF and NB-IoT Comm. module):
- Involvement in deployment of AMI using RF technology, head end system, smart meters and meter data management system and Defining system interface points – interface with SAP ISU and Head End System, Meter Data Management System.
Additional Engagements:
- worked on Quarterly Risk Assessment, calculation of monthly Billing Efficiency, Collection Efficiency and Loss calculation of energy
- Engage in revenue management by implementing analytics on electrical meters data and dealing legal matters of power theft
Education
MBA - Business Management
Indian Institute of Management
Kozhikode, India 04.2001 -
B.Tech - Elec. Engg.
National Institute of Technology - Bhopal
Class XII - Science
Nirmal Happy Senior Secondary School
Class X - undefined
Nehru Public Senior Secondary School
Skills
SQL
undefinedCertification
Analytix-Labs, Data Visualization and exploratory data analysis through advanced excel, SQL and Tableau., Learned and applied various statistical techniques with Python., As a part of training, worked on diff Machine learning projects - Supervised and Unsupervised., Oct'19 to Jan'20
Timeline
Senior Data Scientist
Hero Moto Corp Limited
03.2022 - Current
Assistant Manager
Tata Power Delhi Distribution Limited
06.2018 - 02.2022
Senior Executive
Tata Power Delhi Distribution Limited
07.2016 - 05.2018
MBA - Business Management
Indian Institute of Management
04.2001 -
B.Tech - Elec. Engg.
National Institute of Technology - Bhopal
Class XII - Science
Nirmal Happy Senior Secondary School
Class X - undefined
Nehru Public Senior Secondary School
Analytix-Labs, Data Visualization and exploratory data analysis through advanced excel, SQL and Tableau., Learned and applied various statistical techniques with Python., As a part of training, worked on diff Machine learning projects - Supervised and Unsupervised., Oct'19 to Jan'20
Other Machine Learning Projects on online platform
- Healthcare Analytics (HackathonAnalyticsVidhya): Accurately predict the Length of Stay for each patient to optimize resource allocation and better functioning. (Multinomial Classification), Used Label encoding for categorical variables and performed outlier treatment and Normalization of data., Performed PCA for Feature Engineering, SMOTE for data balancing and XG Boost for modelling., Accuracy : 45% and AV score: 41.92%
- Banking Credit Card Spend Prediction and Identify Drivers for Spend: One of the global banks would like to understand what factors driving credit card spend are. The bank want use these insights to calculate credit limit., Feature Selection Techniques: F-Regression, Recursive Feature Engineering and VIF performed., Established linear regression model on a sample data and reported results with an accuracy of 41%., Implemented machine learning techniques: DecisionTree with ensemble learning, Random forest and Gradient Boosting.
- Indian School's Statistics Analysis: Public Dataset shared by Indian Govt. used from Kaggle comprising boy's and girl's enrolment and drop out ratio at different levels., Exploratory data analysis done using Matplotlib and SeaBorn on different datasets., Significant inferences came out based on the trend i.e. drastic drop in enrolment of girls after secondary
- Walmart Store Sale Prediction: Historical sales data for 45 Walmart stores located in different regions with several promotional markdown events throughout the year., Applied statistical linear Regression model and different ML techniques (RF, GB, and XGB) on a sample data., Finalised the model with average output of statistical and ML models with 16% of MAPE value.
Awards Recognition Scholastic Achievements
- Shining Star For Revenue Protection Module implementation and Logic Framing that benefitted 11Crs to the company.
- Sergeant, NCC Completed C certificate with grade A with the position of Sergeant 2014
- AIEEE Rank Secured All India Rank AIR- 13,313 and was among top1.57% students (Total 10lacs students appeared) 2012
- Merit List 8th rank secured in 12th Board district Merit list 2011
Techniques Used
- Regression Analysis
- Feature Engineering
- Predictive Modelling
- Clustering
- Deep Learning
- Ensemble learning
Other Machine Learning Projects on online platform
- Healthcare Analytics (HackathonAnalyticsVidhya): Accurately predict the Length of Stay for each patient to optimize resource allocation and better functioning. (Multinomial Classification), Used Label encoding for categorical variables and performed outlier treatment and Normalization of data., Performed PCA for Feature Engineering, SMOTE for data balancing and XG Boost for modelling., Accuracy : 45% and AV score: 41.92%
- Banking Credit Card Spend Prediction and Identify Drivers for Spend: One of the global banks would like to understand what factors driving credit card spend are. The bank want use these insights to calculate credit limit., Feature Selection Techniques: F-Regression, Recursive Feature Engineering and VIF performed., Established linear regression model on a sample data and reported results with an accuracy of 41%., Implemented machine learning techniques: DecisionTree with ensemble learning, Random forest and Gradient Boosting.
- Indian School's Statistics Analysis: Public Dataset shared by Indian Govt. used from Kaggle comprising boy's and girl's enrolment and drop out ratio at different levels., Exploratory data analysis done using Matplotlib and SeaBorn on different datasets., Significant inferences came out based on the trend i.e. drastic drop in enrolment of girls after secondary
- Walmart Store Sale Prediction: Historical sales data for 45 Walmart stores located in different regions with several promotional markdown events throughout the year., Applied statistical linear Regression model and different ML techniques (RF, GB, and XGB) on a sample data., Finalised the model with average output of statistical and ML models with 16% of MAPE value.
Awards Recognition Scholastic Achievements
- Shining Star For Revenue Protection Module implementation and Logic Framing that benefitted 11Crs to the company.
- Sergeant, NCC Completed C certificate with grade A with the position of Sergeant 2014
- AIEEE Rank Secured All India Rank AIR- 13,313 and was among top1.57% students (Total 10lacs students appeared) 2012
- Merit List 8th rank secured in 12th Board district Merit list 2011
Techniques Used
- Regression Analysis
- Feature Engineering
- Predictive Modelling
- Clustering
- Deep Learning
- Ensemble learning
Similar Profiles
SAURABH BULANDANISAURABH BULANDANI
AGM - Engine Design & Development, R&D at Hero Moto Corp LimitedAGM - Engine Design & Development, R&D at Hero Moto Corp Limited
AMIT PALAMIT PAL
Assistant General Manager -Central Planning at Hero Moto Corp LimitedAssistant General Manager -Central Planning at Hero Moto Corp Limited
Vaibhav ShrivastavaVaibhav Shrivastava
Manager (LPG)-Sales at Bharat Petroleum Corporation LimitedManager (LPG)-Sales at Bharat Petroleum Corporation Limited